Data Center compatibility

Atlas CRM is officially a "Data Center approved app". After a rigorous technical review process, the app is verified for its ability to perform at a standard suited for Data Center environments.

The Data Center approved program

More information on the Data Center approved program can be found in the Atlassian licensing documentation.

Memory requirements

On installation Atlas CRM requires 256 megabyte of additional ram. This is excluding any memory requirements due to usage scenario's (see below). By default Jira runs with 756 megabytes of ram (-Xmx in Java runtime terms). You should raise this by 256m before installing Atlas CRM.

Here is an example table with other common memory configurations:

Your current configuration

Before installing Atlas CRM

756m

1024m

2048m

2304m

4096m

4352m

8192m

8448m

Storage requirements

Atlas CRM index

Each node requires enough free space in the local home to store the Atlas CRM index. The index grows with the number of entities in the system.

#entities

index size (m is megabytes, g is gigabytes, all by approach)

1000

1m

10000

10m

100000

100m

1000000

1g

For more information on indexing, see below.

Atlas CRM file storage

Additional storage is required to store files that are attached to CRM entities. These files are stored in the shared home. The shared home should be sized according to your storage needs, this greatly depends on your use of Atlas CRM.

Other resource requirements

Depending on your use of the system, you should size your memory and cpu accordingly. There is no general use of thumb for this. It greatly depends on the size of the CRM database and the usage patterns for viewing, searching and editing. Your system sizing should be done according to the system sizing instructions in the Atlassian documentation.

Indexing

All nodes in the cluster build a local index. The indexes are kept in sync by sending cluster messages amongs nodes. When an additional node enters the cluster, it will have to catch up with indexing first before being fully functional in terms of searching and sorting.

Last updated